Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://docs.pytorch.org/docs/stable/generated/torch.unflatten.html below:

torch.unflatten — PyTorch 2.7 documentation

torch.unflatten(input, dim, sizes) → Tensor¶

Expands a dimension of the input tensor over multiple dimensions.

See also

torch.flatten() the inverse of this function. It coalesces several dimensions into one.

Parameters

• input (Tensor) – the input tensor.

• dim (int) – Dimension to be unflattened, specified as an index into input.shape.

• sizes (Tuple[int]) – New shape of the unflattened dimension. One of its elements can be -1 in which case the corresponding output dimension is inferred. Otherwise, the product of sizes must equal input.shape[dim].

Returns

A View of input with the specified dimension unflattened.

Examples::

>>> torch.unflatten(torch.randn(3, 4, 1), 1, (2, 2)).shape
torch.Size([3, 2, 2, 1])
>>> torch.unflatten(torch.randn(3, 4, 1), 1, (-1, 2)).shape
torch.Size([3, 2, 2, 1])
>>> torch.unflatten(torch.randn(5, 12, 3), -2, (2, 2, 3, 1, 1)).shape
torch.Size([5, 2, 2, 3, 1, 1, 3])

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4